Evaluating NLP Systems On a Novel Cloze Task: Judging the Plausibility of Possible Fillers in Instructional Texts
This work addresses the problem of evaluating NLP systems more comprehensively on language understanding for researchers, but it is incremental as it builds on existing cloze tasks and benchmarks.
The paper tackles the limitation of existing cloze tasks by proposing a new task that evaluates NLP systems on judging the plausibility of filler words as good, neutral, or bad candidates, focusing on subtask A in Semeval 2022 task 7 and improving traditional models with an ensemble method.
Cloze task is a widely used task to evaluate an NLP system's language understanding ability. However, most of the existing cloze tasks only require NLP systems to give the relative best prediction for each input data sample, rather than the absolute quality of all possible predictions, in a consistent way across the input domain. Thus a new task is proposed: predicting if a filler word in a cloze task is a good, neutral, or bad candidate. Complicated versions can be extended to predicting more discrete classes or continuous scores. We focus on subtask A in Semeval 2022 task 7, explored some possible architectures to solve this new task, provided a detailed comparison of them, and proposed an ensemble method to improve traditional models in this new task.